Thierry Lafouge

Improving information retrieval by combining user profile and document segmentation

Abstract Due to the ever-increasing quantity of available information, which users have to scan in order to find relevant items, noise has become a major issue in the implementation and use of information retrieval systems. The aim of this study was to design an information retrieval system permitting the “personalization” of search, by taking into account user profile. A pre-orientation system was first developed to give access to a personalized subcorpus. To limit noise in information retrieval systems, the textual material offered to the user is reduced and contains only those sections (units) of the document that interest him and are significant to him (where textual material is used in the sense of document units to be processed by content analysis in order to build descriptions of the documents). In this way, the documents are structured on the basis of utility functions. The selected document units are part of the sub-corpus defined by the pre-orientation system. Next, the profile of each user is characterized by determining competence in a given field and at different levels. Each user is characterized by: • -stable information, related to the person rather than to a particular search. This information provides a general description of the user and his habits, • -variable information, related to a specific search. The priority here is to describe the objective of the search (search may be either exhaustive or non-exhaustive; it may concern specialized or popular publications, etc.). The function of the pre-orientation system is to associate a set of characteristics applying to document units to a given user profile. Search is then applied only to the subset of the selected document units that are relevant to the user and established following his profile. Document units are not characterized on the basis of thematic criteria related to content, but rather on the basis of criteria relating to utility. The objective was to propose a hypothesis on the different parameters determining user profile and document unit characteristics, and to test such a hypothesis using an existing information retrieval system incorporating full-text natural language processing tools.

Comparative analysis between impact factor and h-index for pharmacology and psychiatry journals

Using strictly the same parameters (identical two publication years (2004–2005) and identical one-year citation window (2006)), IF 2006 was compared with h-index 2006 for two samples of “Pharmacology and Pharmacy” and “Psychiatry” journals computed from the ISI Web of Science. For the two samples, the IF and the h-index rankings of the journals are very different. The correlation coefficient between the IF and the h-index is high for Psychiatry but lower for Pharmacology. The linearity test performed between the h-index and

The source-item coverage of the exponential function

\,IF^{{{\frac{\alpha }{\alpha \, + 1}}}} .\,n\,^{{{\frac{1}{\alpha \, + 1}}}}

The customers for document supply in pharmacology: a case study from INIST in France (Part 3)

showed the great sensitivity of the model compared with α. The IF and h-index can be completely complementary when evaluating journals of the same scientific discipline.

Production and use of information. Characterization of informetric distributions using effort function and density function

The study of the statistical regularities observed in the field of information production and use has confirmed the existence of important similarities. Thus, the existence of regularities and measurable ratios allow the prevision and the concept of laws. In the 1950s, Shannon and Weaver [13] modelled the information circulation theory. The entropy hypothesis of this theory is that the more ranked a system is, the less information it produces. Theoretical studies have tried to formalise the connection between the bibliometric distribution and the entropy. In this paper, we try to extend previous results linked with ‘the least effort principle’ and the analytical slope of a bibliometric distribution. In the first and second parts, we recall some statements about entropy and bibliometric distributions and, after that, we describe different links between them.

A new explanation of the geometric law in the case of library circulation data

Statistical distributions in the production of information are most often studied in the framework of Lotkaian informetrics. In this paper, we recall some results of basic theory of Lotkaian informetrics, then we transpose methods (Theorem 1) applied to Lotkaian distributions by Leo Egghe (Theorem 2) to the exponential distributions (Theorem 3, Theorem 4). We give examples and compare the results (Theorem 5). Finally, we propose to widen the problem using the concept of exponential informetric process (Theorem 6).

A cartographic analysis of the correlation between document supply and citations in pharmacology: a case study from INIST in France (part 2)

Purpose: The article investigates the evolution of document supply of print serials. Approach: The study is based on data from INIST-CNRS - document supply requests and access to electronic resources - of 95 serials with JCR impact factors in pharmacology. Data were collected from 1992 to 2003. Findings: The results distinguish four groups of serials each with a different evolution of supply requests. Nevertheless, the overall decline from 1999 is a global phenomenon linked to growing access to online journals and there seems little likelihood of a slowdown for the next few years. Originality: This is the first of three articles on a longitudinal study over 12 years on different aspects of the relationship between document supply and impact factor (citations) in pharmacology, a scientific domain with a high use of scientific information.

An interpretation of the effort function through the mathematical formalism of exponential informetric process

Purpose: The article investigates the customers for the document supply of print serials. Approach: The study is based on data from INIST-CNRS for document supply requests in pharmacology. Data were collected from 1998 to 2005 (10,000+ customer accounts with 4.6m orders). Findings: The article provides information about the distribution of orders per customer type, about the geographical origin of customers, about the ordering trends among pharmaceutical customers and of pharmaceutical journals. Originality: This is the last of three articles on a longitudinal study over 13 years on different aspects of the relationship between document supply and impact factor (citations) in pharmacology, a scientific domain with a high use of scientific information.